Our posers wondered Why XenForo and asked. Most of them are okay with the change from IP.Board now. These are my personal views, and I thought you'd all like it here as well, the same exact thing I told our members:
Glad you guys are enjoying the new features being added slowly... Such as the built in likes and the liking status updates. I can assure you, we're getting a new status update sidebar done! With comments AND the like button for you likeoholics.
I'll take a moment to explain it for others whom may be curious. A lot of the features people loved that were minor features on the old software had huge memory leaks on a large forum such as this. A regular rp forum with only ten to twenty people online would never feel it, but, we were sitting at two hundred people beating on a server with memory leaks from here to there. I'm taking my time adding the new addons for a reason: I'm testing them back/forth for bugs in loading. On IPB with everything on we were using 40.0 server load, on a server with higher specs than we have now. With addons disabled, it dropped down to around 6-8. It was still struggling. We tried the updates, IP.Board Tech, even managed system administrators, nothing worked.
We researched it, and decided on XenForo (thanks to Darkfire for finding introducing it to us) and tested it, read reviews. We researched other large IP.Boards. A lot of the big ones are switching away or are using IP.Board 3.1. Sadly, downgrading from 3.3 to 3.1 was not feasible. When we set up the server with XenForo we did research large XenForo forums as well, with similar resources to us. We attacked their staff and administrators trying to get every detail of how they run their server. It's how we found Ghan whom set our behind the scenes server stuff up, optimized for XenForo. With more average users online than IPB and with the most popular features like the likes button being back on everything, we dropped from a load of 40.0 down to a load of .79 average on half of the resources, HALF.
This is why we switched. If you take the time to look at IP.Board forums and bugs. They're getting frustrating. These problems are being reported in ever increasing quantities. Databases are duplicating themselves due to an IPB bug, we had to ask IPS Tech to fix it a few times. The database is bloating itself. And constant Google Chrome errors, in the posting system, chat, and load. If you open up the archives or another large IP.Board open it in Firefox & Chrome, it hangs big time in Chrome. There are several complaints about it, but, IP.Board isn't focusing on those bug fixes. The major bug fixes that have been in the changelog have mostly been for features for their hosting corporate plans for enterprise. I personally feel that IP.Board is taking the path of vBulletin, catering to the corporations/enterprises now, not people like you and me and communities like ours. vBulletin's high time was in the late 3.X series. I feel IP.Board hit it's peak at 3.1, as it's been downhill since 3.2. Feel free to do the research yourself. IP.Board is amazing for role-plays still, with the features. I hope these bugs get fixed! I recommend staying in the 3.1 series too for now.
I hope this provides insight some of you wanted. I know people keep asking. Now, I'm going to bed.
Edit: Don't quote me on these numbers... As my math brain isn't working. But, rough estimate: .79 average with more online users than IPB's old one. Round to 1.0 to be safe. IPB at 40.0 with doulb ethe resources. So we're technically using what? 2% of what IP.Board used? Edit 3: It's actually closer to 1% of the load cause we halfed the resources needed on the server. But, let's just stick with 2%, I like 2's better.
Edit 2: Oh, before some people ask. I test query speed as well on this new software. I will not let any page get over 1.0 second load time. . Threads have around a .4 load time, boards around .25 and home page is hovering around .8. .4 of the load is that constantly requested addon for collapsing categories. . One addon for example, the statistics advance ajax thing used over .7! Would've drove site to be almost a 2 second load, which is why it didn't make the cut. .